1. https://portal.futuregrid.org Science Applications on Clouds and FutureGrid June 14 2012 Cloud and Autonomic Computing Center Spring 2012 Workshop Cloud Computing: from Cybersecurity to Intercloud University of Florida Gainesville Geoffrey Fox gcf@indiana.edu http://www.infomall.org https://portal.futuregrid.orghttp://www.infomall.orghttps://portal.futuregrid.org Director, Digital Science Center, Pervasive Technology Institute Associate Dean for Research and Graduate Studies, School of Informatics and Computing Indiana University Bloomington (Science Clouds work with Dennis Gannon Microsoft)

2. https://portal.futuregrid.org Science Computing Environments Large Scale Supercomputers – Multicore nodes linked by high performance low latency network – Increasingly with GPU enhancement – Suitable for highly parallel simulations High Throughput Systems such as European Grid Initiative EGI or Open Science Grid OSG typically aimed at pleasingly parallel jobs – Can use “cycle stealing” – Classic example is LHC data analysis Grids federate resources as in EGI/OSG or enable convenient access to multiple backend systems including supercomputers – Portals make access convenient and – Workflow integrates multiple processes into a single job Specialized visualization, shared memory parallelization etc. machines 2

3. https://portal.futuregrid.org Some Observations Classic HPC machines as MPI engines offer highest possible performance on closely coupled problems Clouds offer from different points of view On-demand service (elastic) Economies of scale from sharing Powerful new software models such as MapReduce, which have advantages over classic HPC environments Plenty of jobs making it attractive for students & curricula Security challenges HPC problems running well on clouds have above “cloud” advantages Note 100% utilization of Supercomputers makes elasticity moot for capability (very large) jobs and makes capacity (many modest) use not be on-demand Cloud-HPC Interoperability seems important 3

4. https://portal.futuregrid.org Clouds and Grids/HPC Synchronization/communication Performance Grids > Clouds > Classic HPC Systems Clouds naturally execute effectively Grid workloads but are less clear for closely coupled HPC applications Service Oriented Architectures and workflow appear to work similarly in both grids and clouds May be for immediate future, science supported by a mixture of – Clouds – some practical differences between private and public clouds – size and software – High Throughput Systems (moving to clouds as convenient) – Grids for distributed data and access – Supercomputers (“MPI Engines”) going to exascale

5. https://portal.futuregrid.org What Applications work in Clouds Pleasingly parallel applications of all sorts with roughly independent data or spawning independent simulations – Long tail of science and integration of distributed sensors Commercial and Science Data analytics that can use MapReduce (some of such apps) or its iterative variants (most other data analytics apps) Which science applications are using clouds? – Many demonstrations –Conferences, OOI, HEP …. – Venus-C (Azure in Europe): 27 applications not using Scheduler, Workflow or MapReduce (except roll your own); use worker role – 50% of applications on FutureGrid are from Life Science but there is more computer science than total applications – Locally Lilly corporation is major commercial cloud user (for drug discovery) but Biology department is not 5

6. https://portal.futuregrid.org Parallelism over Users and Usages “Long tail of science” can be an important usage mode of clouds. In some areas like particle physics and astronomy, i.e. “big science”, there are just a few major instruments generating now petascale data driving discovery in a coordinated fashion. In other areas such as genomics and environmental science, there are many “individual” researchers with distributed collection and analysis of data whose total data and processing needs can match the size of big science. Clouds can provide scaling convenient resources for this important aspect of science. Can be map only use of MapReduce if different usages naturally linked e.g. exploring docking of multiple chemicals or alignment of multiple DNA sequences – Collecting together or summarizing multiple “maps” is a simple Reduction 6

7. https://portal.futuregrid.org Internet of Things and the Cloud It is projected that there will be 24 billion devices on the Internet by 2020. Most will be small sensors that send streams of information into the cloud where it will be processed and integrated with other streams and turned into knowledge that will help our lives in a multitude of small and big ways. It is not unreasonable for us to believe that we will each have our own cloud-based personal agent that monitors all of the data about our life and anticipates our needs 24x7. The cloud will become increasing important as a controller of and resource provider for the Internet of Things. As well as today’s use for smart phone and gaming console support, “smart homes” and “ubiquitous cities” build on this vision and we could expect a growth in cloud supported/controlled robotics. Natural parallelism over “things” 7

8. https://portal.futuregrid.org Sensors as a Service Sensor Processing as a Service (could use MapReduce) A larger sensor ……… Output Sensor

9. https://portal.futuregrid.org Classic Parallel Computing HPC: Typically SPMD (Single Program Multiple Data) “maps” typically processing particles or mesh points interspersed with multitude of low latency messages supported by specialized networks such as Infiniband and technologies like MPI – Often run large capability jobs with 100K cores on same job – National DoE/NSF/NASA facilities run 100% utilization – Fault fragile and cannot tolerate “outlier maps” taking longer than others Clouds: MapReduce has asynchronous maps typically processing data points with results saved to disk. Final reduce phase integrates results from different maps – Fault tolerant and does not require map synchronization – Map only useful special case HPC + Clouds: Iterative MapReduce caches results between “MapReduce” steps and supports SPMD parallel computing with large messages as seen in parallel linear algebra need in clustering and other data mining 9

10. https://portal.futuregrid.org 4 Forms of MapReduce 10

11. https://portal.futuregrid.org Number of Executing Map Task Histogram Strong Scaling with 128M Data Points Weak Scaling Task Execution Time Histogram First iteration performs the initial data fetch Overhead between iterations Scales better than Hadoop on bare metal

12. https://portal.futuregrid.org Data Intensive Iterative Applications I Important class of (Data analytics) applications – Driven by data deluge & emerging fields – Data mining, machine learning – often with linear algebra at core – Expectation maximization k ← 0; MAX ← maximum iterations δ [0] ← initial paramwter value while ( k< MAX_ITER || f(δ [k], δ [k-1] ) ) foreach datum in data β[datum] ← process (datum, δ [k] ) end foreach δ [k+1] ← combine(β[]) k ← k+1 end while k ← 0; MAX ← maximum iterations δ [0] ← initial paramwter value while ( k< MAX_ITER || f(δ [k], δ [k-1] ) ) foreach datum in data β[datum] ← process (datum, δ [k] ) end foreach δ [k+1] ← combine(β[]) k ← k+1 end while

13. https://portal.futuregrid.org Twister for Data Intensive Iterative Applications (Iterative) MapReduce structure Iterative Map-Collective is framework Twister runs on Linux or Azure Twister4Azure is built on top of Azure tables, queues, storage Judy Qiu IU Compute CommunicationReduce/ barrier New Iteration Larger Loop- Invariant Data Generalize to arbitrary Collective Broadcast Smaller Loop- Variant Data

14. https://portal.futuregrid.org What to use in Clouds HDFS style file system to collocate data and computing Queues to manage multiple tasks Tables to track job information MapReduce and Iterative MapReduce to support parallelism Services for everything Portals as User Interface Appliances and Roles (Venus-C approach) as customized images Software environments/tools like Google App Engine, memcached Workflow to link multiple services (functions) 14

15. https://portal.futuregrid.org What to use in Grids and Supercomputers? Portals and Workflow as in clouds MPI and GPU/multicore threaded parallelism Services in Grids Wonderful libraries supporting parallel linear algebra, particle evolution, partial differential equation solution Parallel I/O for high performance in an application Wide area File System (e.g. Lustre) supporting file sharing This is a rather different style of PaaS from clouds – should we unify? 15

16. https://portal.futuregrid.org How to use Clouds I 1)Build the application as a service. Because you are deploying one or more full virtual machines and because clouds are designed to host web services, you want your application to support multiple users or, at least, a sequence of multiple executions. If you are not using the application, scale down the number of servers and scale up with demand. Attempting to deploy 100 VMs to run a program that executes for 10 minutes is a waste of resources because the deployment may take more than 10 minutes. To minimize start up time one needs to have services running continuously ready to process the incoming demand. 2)Build on existing cloud deployments. For example use an existing MapReduce deployment such as Hadoop or existing Roles and Appliances (Images) 16

17. https://portal.futuregrid.org How to use Clouds II 3)Use PaaS if possible. For platform-as-a-service clouds like Azure use the tools that are provided such as queues, web and worker roles and blob, table and SQL storage. 3)Note HPC systems don’t offer much in PaaS area 4)Design for failure. Applications that are services that run forever will experience failures. The cloud has mechanisms that automatically recover lost resources, but the application needs to be designed to be fault tolerant. In particular, environments like MapReduce (Hadoop, Daytona, Twister4Azure) will automatically recover many explicit failures and adopt scheduling strategies that recover performance "failures" from for example delayed tasks. One expects an increasing number of such Platform features to be offered by clouds and users will still need to program in a fashion that allows task failures but be rewarded by environments that transparently cope with these failures. (Need to build more such robust environments) 17

18. https://portal.futuregrid.org How to use Clouds III 5)Use as a Service where possible. Capabilities such as SQLaaS (database as a service or a database appliance) provide a friendlier approach than the traditional non-cloud approach exemplified by installing MySQL on the local disk. Suggest that many prepackaged aaS capabilities such as Workflow as a Service for eScience will be developed and simplify the development of sophisticated applications. 6)Moving Data is a challenge. The general rule is that one should move computation to the data, but if the only computational resource available is a the cloud, you are stuck if the data is not also there. Persuade Cloud Vendor to host your data free in cloud Persuade Internet2 to provide good link to Cloud Decide on Object Store v. HDFS style (or v. Lustre WAFS on HPC) 18

19. https://portal.futuregrid.org Architecture of Data Repositories? Traditionally governments set up repositories for data associated with particular missions – For example EOSDIS (Earth Observation), GenBank (Genomics), NSIDC (Polar science), IPAC (Infrared astronomy) – LHC/OSG computing grids for particle physics This is complicated by volume of data deluge, distributed instruments as in gene sequencers (maybe centralize?) and need for intense computing like Blast – i.e. repositories need lots of computing? 19

20. https://portal.futuregrid.org Clouds as Support for Data Repositories? The data deluge needs cost effective computing – Clouds are by definition cheapest – Need data and computing co-located Shared resources essential (to be cost effective and large) – Can’t have every scientists downloading petabytes to personal cluster Need to reconcile distributed (initial source of ) data with shared analysis – Can move data to (discipline specific) clouds – How do you deal with multi-disciplinary studies Data repositories of future will have cheap data and elastic cloud analysis support? – Hosted free if data can be used commercially? 20

21. https://portal.futuregrid.org Using Science Clouds in a Nutshell High Throughput Computing; pleasingly parallel; grid applications Multiple users (long tail of science) and usages (parameter searches) Internet of Things (Sensor nets) as in cloud support of smart phones (Iterative) MapReduce including “most” data analysis Exploiting elasticity and platforms (HDFS, Queues..) Use worker roles, services, portals (gateways) and workflow Good Strategies: – Build the application as a service; – Build on existing cloud deployments such as Hadoop; – Use PaaS if possible; – Design for failure; – Use as a Service (e.g. SQLaaS) where possible; – Address Challenge of Moving Data 21

22. https://portal.futuregrid.org FutureGrid key Concepts I FutureGrid is an international testbed modeled on Grid5000 Supporting international Computer Science and Computational Science research in cloud, grid and parallel computing (HPC) – Industry and Academia The FutureGrid testbed provides to its users: – A flexible development and testing platform for middleware and application users looking at interoperability, functionality, performance or evaluation – Each use of FutureGrid is an experiment that is reproducible – A rich education and teaching platform for advanced cyberinfrastructure (computer science) classes

23. https://portal.futuregrid.org FutureGrid key Concepts II FutureGrid has a complementary focus to both the Open Science Grid and the other parts of XSEDE (TeraGrid). – FutureGrid is user-customizable, accessed interactively and supports Grid, Cloud and HPC software with and without virtualization. – FutureGrid is an experimental platform where computer science applications can explore many facets of distributed systems – and where domain sciences can explore various deployment scenarios and tuning parameters and in the future possibly migrate to the large-scale national Cyberinfrastructure. – FutureGrid supports Interoperability Testbeds (see OGF) Note much of current use Education, Computer Science Systems and Biology/Bioinformatics

24. https://portal.futuregrid.org FutureGrid key Concepts III Rather than loading images onto VM’s, FutureGrid supports Cloud, Grid and Parallel computing environments by provisioning software as needed onto “bare-metal” using Moab/xCAT –Image library for MPI, OpenMP, MapReduce (Hadoop, Dryad, Twister), gLite, Unicore, Globus, Xen, ScaleMP (distributed Shared Memory), Nimbus, Eucalyptus, OpenNebula, KVM, Windows ….. –Either statically or dynamically Growth comes from users depositing novel images in library FutureGrid has ~4500 distributed cores with a dedicated network and a Spirent XGEM network fault and delay generator Image1 Image2 ImageN … LoadChooseRun

25. https://portal.futuregrid.org FutureGrid Partners Indiana University (Architecture, core software, Support) San Diego Supercomputer Center at University of California San Diego (INCA, Monitoring) University of Chicago/Argonne National Labs (Nimbus) University of Florida (ViNE, Education and Outreach, Operations) University of Southern California Information Sciences (Pegasus for workflow and to manage experiments) University of Tennessee Knoxville (Benchmarking) University of Texas at Austin/Texas Advanced Computing Center (Portal) University of Virginia (OGF, XSEDE) Center for Information Services and GWT-TUD from Technische Universtität Dresden. (VAMPIR) Red institutions have FutureGrid hardware

26. https://portal.futuregrid.org FutureGrid: a Grid/Cloud/HPC Testbed Private Public FG Network NID : Network Impairment Device

27. https://portal.futuregrid.org Compute Hardware NameSystem type# CPUs# CoresTFLOPS Total RAM (GB) Secondary Storage (TB) Site Status india IBM iDataPlex2561024113072339 + 16IU Operational alamo Dell PowerEdge 1927688115230TACC Operational hotel IBM iDataPlex16867272016120UC Operational sierra IBM iDataPlex1686727268896SDSC Operational xray Cray XT5m16867261344339IU Operational foxtrot IBM iDataPlex64256276824UF Operational Bravo Large Disk & memory 321281.5 3072 (192GB per node) 144 (12 TB per Server) IU Operational Delta Large Disk & memory With Tesla GPU’s 32 32 GPU’s 192+ 14336 GPU ? 6 1536 (192GB per node) 96 (12 TB per Server)IUOperational TOTAL Cores 4384

28. https://portal.futuregrid.org Storage Hardware System TypeCapacity (TB)File SystemSiteStatus Xanadu 360180NFSIUNew System DDN 6620120GPFSUCNew System SunFire x417096ZFSSDSCNew System Dell MD300030NFSTACCNew System IBM24NFSUFNew System Substantial back up storage at IU: Data Capacitor and HPSS

29. https://portal.futuregrid.org Network Impairment Device Spirent XGEM Network Impairments Simulator for jitter, errors, delay, etc Full Bidirectional 10G w/64 byte packets up to 15 seconds introduced delay (in 16ns increments) 0-100% introduced packet loss in.0001% increments Packet manipulation in first 2000 bytes up to 16k frame size TCL for scripting, HTML for manual configuration

30. https://portal.futuregrid.org FutureGrid: Inca Monitoring

31. https://portal.futuregrid.org 5 Use Types for FutureGrid 218 approved projects June 13 2012 – https://portal.futuregrid.org/projects Training Education and Outreach (8%) – Semester and short events; promising for small universities Interoperability test-beds (3%) – Grids and Clouds; Standards; from Open Grid Forum OGF Domain Science applications (31%) – Life science highlighted (18%), Non Life Science (13%) Computer science (47%) – Largest current category Computer Systems Evaluation (27%) – XSEDE (TIS, TAS), OSG, EGI Clouds are meant to need less support than other models; FutureGrid needs more user support ……. 31

32. https://portal.futuregrid.org 32 https://portal.futuregrid.org/projects

33. https://portal.futuregrid.org Recent Projects 33

34. https://portal.futuregrid.org Distribution of FutureGrid Technologies and Areas 220 Projects

35. https://portal.futuregrid.org Environments Chosen Fractions v Time 35 High Performance Computing Environment: 103(47.2%) Eucalyptus: 109(50%) Nimbus: 118(54.1%) Hadoop: 75(34.4%) Twister: 34(15.6%) MapReduce: 69(31.7%) OpenNebula: 32(14.7%) OpenStack: 36(16.5%) Genesis II: 29(13.3%) XSEDE Software Stack: 44(20.2%) Unicore 6: 16(7.3%) gLite: 17(7.8%) Vampir: 10(4.6%) Globus: 13(6%) Pegasus: 10(4.6%) PAPI: 8(3.7%) CUDA(GPU Software)): 1(0.5%)

36. https://portal.futuregrid.org FutureGrid Tutorials Cloud Provisioning Platforms Using Nimbus on FutureGrid [novice] Nimbus One-click Cluster Guide Using OpenStack Nova on FutureGrid Using Eucalyptus on FutureGrid [novice] Connecting private network VMs across Nimbus clusters using ViNe [novice] Using the Grid Appliance to run FutureGrid Cloud Clients [novice] Cloud Run-time Platforms Running Hadoop as a batch job using MyHadoop [novice] Running SalsaHadoop (one-click Hadoop) on HPC environment [beginner] Running Twister on HPC environment Running SalsaHadoop on Eucalyptus Running FG-Twister on Eucalyptus Running One-click Hadoop WordCount on Eucalyptus [beginner] Running One-click Twister K-means on Eucalyptus Image Management and Rain Using Image Management and Rain [novice] Storage Using HPSS from FutureGrid [novice] Educational Grid Virtual Appliances Running a Grid Appliance on your desktop Running a Grid Appliance on FutureGrid Running an OpenStack virtual appliance on FutureGrid Running Condor tasks on the Grid Appliance Running MPI tasks on the Grid Appliance Running Hadoop tasks on the Grid Appliance Deploying virtual private Grid Appliance clusters using Nimbus Building an educational appliance from Ubuntu 10.04 Customizing and registering Grid Appliance images using Eucalyptus High Performance Computing Basic High Performance Computing Running Hadoop as a batch job using MyHadoop Performance Analysis with Vampir Instrumentation and tracing with VampirTrace Experiment Management Running interactive experiments [novice] Running workflow experiments using Pegasus Pegasus 4.0 on FutureGrid Walkthrough [novice] Pegasus 4.0 on FutureGrid Tutorial [intermediary] Pegasus 4.0 on FutureGrid Virtual Cluster [advanced] 36

37. https://portal.futuregrid.org Software Components Portals including “Support” “use FutureGrid” “Outreach” Monitoring – INCA, Power (GreenIT) Experiment Manager: specify/workflow Image Generation and Repository Intercloud Networking ViNE Virtual Clusters built with virtual networks Performance library Rain or Runtime Adaptable InsertioN Service for images Security Authentication, Authorization, Note Software integrated across institutions and between middleware and systems Management (Google docs, Jira, Mediawiki) Note many software groups are also FG users “Research” Above and below Nimbus OpenStack Eucalyptus

38. https://portal.futuregrid.org Motivation: Image Management and Rain on FutureGrid Allow users to take control of installing the OS on a system on bare metal (without the administrator) By providing users with the ability to create their own environments to run their projects (OS, packages, software) Users can deploy their environments in both bare-metal and virtualized infrastructures Security is obviously important RAIN manages tools to dynamically provide custom HPC environment, Cloud environment, or virtual networks on- demand http://futuregrid.org

39. https://portal.futuregrid.org Architecture

40. https://portal.futuregrid.org Create Image from Scratch CentOS Ubuntu

41. https://portal.futuregrid.org Create Image from Base Image CentOS Ubuntu

42. https://portal.futuregrid.org Templated Dynamic Provisioning 42 Abstract Specification of image mapped to various HPC and Cloud environments Essex replaces Cactus Current Eucalyptus 3 commercial while version 2 Open Source OpenNebula Parallel provisioning now supported Moab/xCAT HPC – high as need reboot before use

43. https://portal.futuregrid.org What is FutureGrid? The FutureGrid project mission is to enable experimental work that advances: a)Innovation and scientific understanding of distributed computing and parallel computing paradigms, b)The engineering science of middleware that enables these paradigms, c)The use and drivers of these paradigms by important applications, and, d)The education of a new generation of students and workforce on the use of these paradigms and their applications. The implementation of mission includes Distributed flexible hardware with supported use Identified IaaS and PaaS “core” software with supported use Outreach ~4500 cores in 5 major sites FutureGrid Usage